With an increase in the elderly population, there is a growing interest in the effects of androgen replacement therapy in hypogonadal older men. Male hypogonadism is associated with a decline in libido, lean body mass, muscle strength, bone density, and sexual function, along with changes in mood and cognition. Unlike menopause in women, where there is an absolute estrogen deficiency, andropause in men is characterized by the gradual decline in biologically active androgen levels. Many elderly men maintain free androgen levels that are in the low normal range. Commonly, if a physician determines that testosterone replacement is needed, intra-muscular testosterone injections, patches or gels are prescribed. In this study we will examine the possibility of increasing testosterone levels in another way, i.e., by causing the testes to secrete more testosterone by blocking estrogen synthesis. A young man normally has testosterone levels ranging between 350-1100 ng/dl. About 20% of men over the age of 70 years have serum testosterone levels of 350 ng/dl or less and are, therefore, considered testosterone deficient. The deficiency can result in osteoporosis, sarcopenia, lack of libido and may contribute to cognitive decline. Achieving serum testosterone levels of 500-550 ng/dl is the therapeutic goal. Experimental data suggest that the GnRH (gonadotropin releasing hormone) pulse generator in the hypothalamus becomes inactive in elderly men who have low testosterone levels and this deficiency results in lower gonadotropin secretion from the pituitary with the resultant lower testosterone secretion from the end organ, the testicles. The gonadotropins, LH and FSH, are secreted in a pulsatile manner and in elderly men with low plasma testosterone levels, the pulses are severely blunted, again pointing to defective drive by the GnRH pulse generator. Elderly subjects given exogenous GnRH do, in fact, have a gonadotropin response as robust (and sometimes more robust) than younger men, again concurring with the evidence that there is a deficient hypothalamic GnRH pulse generator. As indicated, men over the age of 70 years may be testosterone deficient. We propose to recruit men ages 65 years and older because testosterone levels in men start declining from the age of 40. Approximately 20% of men above the age of 65 or 70 yrs (based on various studies) are hypogonadal, based on serum testosterone levels of 300 mg/dl or less. This increases to 50% in men >80 yrs old. The reason for selecting the cut-off age of 65 years is based on the hypothesis that earlier intervention in these hypogonadal men will prevent the development of frailty syndromes like loss of muscle mass, muscle strength, bone mass and other cognitive and sexual function. As outlined in the recruitment strategy, we will select only those men who have testosterone levels <350 ng/dl (which is the lower limit of normal in young men). Since all men 65 and above are eligible to participate in this study, men in their 7th or 8th decade of life, if they meet the selection criteria, will be part of the study. In men, both types of estrogens, estrone and estradiol, are present. Most male estrogen arises from peripheral conversion of testosterone and other androgenic precursors by aromatase activity in fat. A substantial fraction of androgen precursors are derived from the adrenal glands. Therefore, deficiency of testosterone secretion can cause reductions in circulating androgens with much less effect on circulating estrogens. This results in an increase in the estrogen/androgen ratio. This altered ratio is the likely cause of mild gynecomastia that is sometimes seen in elderly men. Giving exogenous testosterone will also result in raising estrogen levels due to the conversion of a fraction of testosterone to estrogen. The protection of bones and decreased libido are the primary indication for testosterone replacement in elderly testosterone deficient men. Both androgens and estrogens contribute to bone health in men. An alternative approach to increasing circulating testosterone levels, other than exogenous administration, is to prevent conversion of circulating testosterone to estrogen by preventing its aromatization, that is, by using an aromatase inhibitor such as anastrozole. As a result, deficiency of estrogen, in turn, activates the GnRH pulse generator due to lack of a negative feedback. This increase in GnRH secretion leads to stimulation of the gonadotropins, which, in turn stimulate the testicles to produce more testosterone. Men given an aromatase inhibitor in a 9-week study had an increase in testosterone levels of about 57%. Studies in men with aromatase inhibition have always been short term. We do not know, for example, if aromatase inhibition is lost if aromatase inhibitors are used over a long period. Use of an aromatase inhibitor to elevate serum testosterone will not raise estrogen levels. Comparatively, exogenous testosterone will raise estrogen levels, therefore, giving us an opportunity to separate the physiological effects of testosterone per se from the estrogenic effects. An added benefit of this study is that we will also find out if exogenous testosterone replacement has beneficial effects on bone markers, lipid levels, indices of metabolism and well-being. Anastrozole is a non-steroidal aromatase inhibitor which is FDA approved for treatment of breast cancer. It will decrease peripheral estrogen levels, and this lower level will stimulate the hypothalamus to increase GnRH. This stimulus in turn, will drive the pituitary gland to release LH/FSH, which will result in increased testosterone release from the Leydig cells of the testis. We expect that the increased testosterone levels secondary to anastrozole should have positive outcomes such as, increased sense of well-being, improved sexual function and increased muscle strength. As anastrozole decreases plasma estrogen levels, we will examine whether this change influences bone density and markers of bone turnover. It is conceivable that estrogen levels after being low for a short term, ultimately recover to baseline. Hence, there is a need for a study of at least one year duration to evaluate the efficacy of aromatase inhibitors in maintaining their inhibition. The sex hormones, both estrogen and testosterone, are known to influence insulin resistance. Testosterone administration to hypogonadal men improves insulin resistance. On the other hand, increased endogenous testosterone production (e.g., women with polycystic ovary syndrome) or testosterone administration at high doses in women result in glucose intolerance and even frank diabetes. We do not know the role of estrogen on glucose tolerance in men. Also, unclear is the fact that the improvement in insulin resistance in men when given testosterone is due to the direct effects of testosterone or due to its aromatization to estrogen. We plan to answer this question with this study.